Leukemia is the most common cancer among children and a significant cause of morbidity, stress, and long-term sequelae. Despite advances in treatment that permit a cure in the majority of leukemias, we still know very little about the causes of the disease, thwarting any attempt at prevention. Our focus is to describe in molecular detail the chromosomal and mutational changes that occur in leukemias, and use these genetic aberrations to develop tools to study the timing and causality of these events. Hitherto we have focused on translocations, finding that most but not all translocation subtypes have an in utero origin. The most common childhood translocation, t(12;21) TEL-AML1, occurs in 25% of childhood acute lymphocytic leukemia and usually, if not always, occurs before birth in children who later contract leukemia with this genetic aberration. We now propose to describe the secondary genetic events in TEL-AMLI+ leukemia, focusing first on the 12p deletion, which is a common (~75%) event in TEL-AMLI+ leukemia. We will use array-Comparative Genomic Hybridization (array-CGH) to define the exact breakpoints of 12p chromosomal deletions, allowing their cloning at the genomic nucleotide level. Next, we will use the cloned fusion junction as a probe to "backtrack" leukemia to birth, by searching for the presence of this genetic marker on neonatal heel-prick Guthrie cards from the same children that we have also sequenced and "backtracked" the genomic TEL-AML 1 translocation sequence. Thirdly, we will use 12p deletions along with TEL-AML 1 translocations to track the fate of the leukemia clone in children after therapy, thereby completing the story of the "natural history" of these genomic fusions and determining clinical utility of the novel markers. Lastly, we will assess the presence of putative tumor suppressor allele(s) that are indicated as allelic loss in array-CGH experiments. Preliminary results suggest the presence of more than one distinct deletion on 12p in some patients, demonstrating the complexity and discovery potential of this analysis. Our results will inform and guide the Northern California Childhood Leukemia Study, an epidemiological project from which all of our samples are derived. We are ultimately working to elucidate the origin and causes of a cancer that might be preventable. [unreadable] [unreadable]